COVID-19 Progression: Australian Data Only

library(tidyverse)
library(rlang)
library(lubridate)
library(scales)
library(ggrepel)
library(glue)
library(rvest)
library(pander)
library(plotly)
library(QuantTools)
library(jsonlite)
panderOptions("big.mark", ",")
panderOptions("table.split.table", Inf)
panderOptions("table.style", "rmarkdown")
panderOptions("missing", "")
theme_set(theme_bw())

auStates <- c(
  ACT = "Australian Capital Territory",
  QLD = "Queensland",
  NSW = "New South Wales",
  VIC = "Victoria",
  SA = "South Australia",
  WA = "Western Australia",
  NT = "Northern Territory",
  TAS = "Tasmania"
)

ausPops <- tribble(
  ~State, ~Population,
  "New South Wales",    8117976,
  "Victoria", 6629870,
  "Queensland", 5115451,
  "South Australia", 1756494,
  "Western Australia", 2630557,
  "Tasmania", 535500,
  "Northern Territory", 245562,
  "Australian Capital Territory", 428060
)

Data Sources

data <- fromJSON("https://covidlive.com.au/covid-live.json") %>%
    as_tibble() %>%
    mutate(
      across(
        .cols = ends_with("CNT"), 
        .fns = as.numeric
      ),
      REPORT_DATE = ymd(REPORT_DATE)
    )

dt <- data %>%
    dplyr::filter(
        CODE == "AUS", !is.na(LAST_UPDATED_DATE)
    ) %>%
    pull(REPORT_DATE) %>%
    max()

getCovidTable <- function(
  state = c("ACT", "QLD", "NSW", "VIC", "SA", "WA", "NT", "TAS"),
  type = c("cases", "recoveries", "deaths", "active-cases", "tests"),
  .state_full = auStates
){
  
  state <- match.arg(state)
  
  type <- match.arg(type)
  type_name <- c(
    cases = "confirmed",
    recoveries = "recovered",
    deaths = "deaths",
    `active-cases` = "active",
    tests = "tests"
  )[[type]]
  col_name <- c(
    cases = "CASES",
    recoveries = "RECOV",
    deaths = "DEATHS",
    `active-cases` = "ACTIVE",
    tests = "TESTS"
  )[[type]]

    url <- glue("https://covidlive.com.au/report/daily-{type}/{str_to_lower(state)}")
  url %>%
    read_html() %>%
    html_nodes("body") %>%
    xml_find_all(
      glue("//table[contains(@class, 'DAILY-{str_to_upper(type)}')]")
    ) %>% 
    html_table() %>%
    .[[1]] %>%
    as_tibble() %>%
    dplyr::select(date = DATE, !!sym(type_name) := !!sym(col_name)) %>%
    separate(date, into = c("day", "month")) %>%
    mutate(
      year = case_when(
        month %in% c("Jan") & day < 25 ~ 2021L,
        TRUE ~ 2020L
      ),
      date = paste(year, month, day, sep = "-"),
      date = parse_date_time(date, orders = "%Y-%B-%d"),
      date = ymd(date),
      State = .state_full[str_to_upper(state)],
      Country = "Australia"
    ) %>%
    mutate_at(
      .vars = type_name, 
      .funs = function(x){as.integer(str_remove_all(x, ","))}
    ) %>%
    dplyr::select(State, Country, date, !!sym(type_name)) %>%
    arrange(date) 
}

confirmed <- names(auStates) %>%
  lapply(getCovidTable, type = "cases") %>%
  bind_rows()

recovered <- names(auStates) %>%
  lapply(getCovidTable, type = "recoveries") %>%
  bind_rows()

deaths <- names(auStates) %>%
  lapply(getCovidTable, type = "deaths") %>%
  bind_rows()

active <- names(auStates) %>%
  lapply(getCovidTable, type = "active-cases") %>%
  bind_rows()

tested <- names(auStates) %>%
  lapply(getCovidTable, type = "tests") %>%
  bind_rows() %>%
  dplyr::filter(!is.na(tests))

getLocal <- function(  
  state = c("ACT", "QLD", "NSW", "VIC", "SA", "WA", "NT", "TAS"),  
  .state_full = auStates
){
  
  state <- match.arg(state)
  
  url <- glue("https://covidlive.com.au/report/daily-source-overseas/{str_to_lower(state)}")
  url %>%
    read_html() %>%
    html_nodes("body") %>%
    xml_find_all("//table[contains(@class, 'DAILY-SOURCE-OVERSEAS')]") %>% 
    html_table() %>%
    .[[1]] %>%
    as_tibble() %>%
    dplyr::select(date = DATE, local = LOCAL) %>%
    separate(date, into = c("day", "month")) %>%
    mutate(
      year = case_when(
        month %in% c("Jan") & day < 25 ~ 2021L,
        TRUE ~ 2020L
      ),
      date = paste(year, month, day, sep = "-"),
      date = parse_date_time(date, orders = "%Y-%B-%d"),
      date = ymd(date),
      State = .state_full[str_to_upper(state)],
      Country = "Australia",
      local = str_remove_all(local, ","),
      local = as.integer(local)
    ) %>%
    dplyr::select(State, Country, date, local) %>%
    arrange(date) 
}
local <- names(auStates) %>%
  lapply(getLocal) %>%
  bind_rows() %>%
  dplyr::filter(!is.na(local))

vic_aged <- "https://covidlive.com.au/report/daily-outbreaks-aged" %>%
  read_html() %>%
  html_nodes("body") %>%
  xml_find_all("//table[contains(@class, 'DAILY-OUTBREAKS-AGED')]") %>% 
    html_table() %>%
    .[[1]] %>%
    as_tibble() %>%
    dplyr::select(date = DATE, deaths = DEATHS, active = ACTIVE) %>%
    separate(date, into = c("day", "month")) %>%
    mutate(
      year = case_when(
        month %in% c("Jan") & day < 25 ~ 2021L,
        TRUE ~ 2020L
      ),
      date = paste(year, month, day, sep = "-"),
      date = parse_date_time(date, orders = "%Y-%b-%d"),
      date = ymd(date),
      State = "Victoria (Aged Care)",
      Country = "Australia"
    ) %>%
  arrange(date) %>%
  mutate(
    deaths = na_locf(deaths),
    active = na_locf(active),
    new = c(active[[1]], diff(active)),
    new = ifelse(new < 0, 0, new),
    confirmed = cumsum(new),
    recovered = confirmed - deaths - active,
    recovered = ifelse(recovered < 0, 0, recovered)
  ) %>%
  dplyr::filter(date <= dt) %>%
  dplyr::select(
    State, Country, date, deaths, confirmed, active, recovered
  ) 

Data for confirmed cases, active cases, recoveries and fatalities was exclusively sourced from COVID LIVE. Reliable data sources were extremely challenging prior to April and as such, minimal data is available from the early stages of the outbreak.

Similarly, data for the Victorian Aged Care outbreak was obtained from data based on press releases.

International Data

International data and figures can be viewed here

Latest Australian Data

Australian State populations were taken from the ABS Website and were accurate in Sept 2019.

• Using an estimated population size of 25,459,470, the total percentage of the Australian population confirmed as having been infected currently sits at 0.11%, or one person in every 889.
• Within Victoria, that rises to one in every 325 having contracted the virus at some point

data %>%
    dplyr::filter(
        REPORT_DATE == dt
    ) %>%
    dplyr::mutate(
        Increase = CASE_CNT - PREV_CASE_CNT,
        `% Increase` = percent(Increase / PREV_CASE_CNT, accuracy = 0.1),
        `Fatality Rate` = percent(DEATH_CNT / CASE_CNT, accuracy = 0.1),
        `Recovery Rate` = percent(RECOV_CNT / CASE_CNT, accuracy = 0.1),
        State = case_when(
            CODE == "AUS" ~ "National Total",
            TRUE ~ auStates[CODE]
        ),
        State = factor(State, levels = c(dplyr::arrange(ausPops, desc(Population))$State, "National Total"))
    ) %>%
    dplyr::rename(
        Fatalities = DEATH_CNT,
        Recovered = RECOV_CNT,
        `Currently Active` = ACTIVE_CNT
    ) %>%
    dplyr::select(
        State,
        PREV_CASE_CNT, CASE_CNT,
        contains("Increase"),
        contains("Fatal"),
        contains("Recov", ignore.case = FALSE),
        `Currently Active`
    ) %>%
    dplyr::arrange(State) %>%
    setNames(
        str_replace_all(names(.), "PREV_CASE_CNT", as.character(dt - 1))
    ) %>%
    setNames(
        str_replace_all(names(.), "CASE_CNT", as.character(dt))
    ) %>%
    pander(
        justify = "lrrrrrrrrr",
        caption = paste(
            "*Confirmed cases, fatalities and recoveries reported by each state at time of preparation.*"
        ),
        emphasize.strong.rows = nrow(.)
    )

Confirmed cases, fatalities and recoveries reported by each state at time of preparation.

State	2021-01-11	2021-01-12	Increase	% Increase	Fatalities	Fatality Rate	Recovered	Recovery Rate	Currently Active
New South Wales	5,018	5,034	16	0.3%	54	1.1%	3,216	63.9%	200
Victoria	20,411	20,411	0	0.0%	820	4.0%	19,553	95.8%	38
Queensland	1,278	1,281	3	0.2%	6	0.5%	1,243	97.0%	26
Western Australia	877	878	1	0.1%	9	1.0%	855	97.4%	14
South Australia	588	588	0	0.0%	4	0.7%	569	96.8%	15
Tasmania	234	234	0	0.0%	13	5.6%	221	94.4%	0
Australian Capital Territory	118	118	0	0.0%	3	2.5%	115	97.5%	0
Northern Territory	90	90	0	0.0%	0	0.0%	71	78.9%	19
National Total	28,614	28,634	20	0.1%	909	3.2%	25,843	90.3%	312

Plot of Current Australian Values

ausStatsCap <- "*Current confirmed and recovered cases, along with fatalities for Australia only. Active cases are shown as confirmed cases excluding fatalities and those classed as recovered. Some data regarding recovered cases prior to 1^st^ May 2020 may be estimates.*"

ggplotly(
  data %>% 
    dplyr::filter(CODE == "AUS", REPORT_DATE <= dt, REPORT_DATE > "2020-03-01") %>% 
    dplyr::select(
      Date = REPORT_DATE, 
      Confirmed = CASE_CNT,
      Active = ACTIVE_CNT, 
      Fatal = DEATH_CNT, 
      Recovered = RECOV_CNT
    ) %>% 
    mutate(
      Active = case_when(
        is.na(Active) ~ Confirmed - Fatal - Recovered,
        TRUE ~ Active
      )
    )  %>%
    pivot_longer(
      cols = c("Active", "Fatal", "Recovered"), 
      names_to = "Status", values_to = "Total"
    ) %>%
    mutate(
      Status = factor(Status, levels = c("Fatal", "Recovered", "Active"))
    ) %>%
    ggplot(aes(Date, Total, fill = Status)) +
    geom_col() +
    geom_line(
      data = . %>%
        group_by(Date) %>%
        summarise(
          Total = sum(Total)
        ) %>%
        mutate(Status = "Confirmed"),
      colour = "blue"
    ) +
    scale_fill_manual(
      values = c(
        Active = rgb(0, 0, 0),
        Confirmed = rgb(0, 0.3, 0.7),
        Fatal = rgb(0.8, 0.2, 0.2),
        Recovered = rgb(0.2, 0.7, 0.4)
      )
    ) +
    scale_x_date(expand = expansion(c(0, 0.03))) +
    scale_y_continuous(expand = expansion(c(0, 0.05))) +
    labs("Total Cases")
)

Current confirmed and recovered cases, along with fatalities for Australia only. Active cases are shown as confirmed cases excluding fatalities and those classed as recovered. Some data regarding recovered cases prior to 1^st May 2020 may be estimates.

ggplotly(
  data %>% 
    dplyr::filter(CODE != "AUS", REPORT_DATE <= dt, REPORT_DATE > "2020-03-01") %>% 
    dplyr::select(
      CODE,
      Date = REPORT_DATE, 
      Confirmed = CASE_CNT,
      Active = ACTIVE_CNT, 
      Fatal = DEATH_CNT, 
      Recovered = RECOV_CNT
    ) %>% 
    mutate(
      Active = case_when(
        is.na(Active) ~ Confirmed - Fatal - Recovered,
        TRUE ~ Active
      ),
      State = auStates[CODE]
    )  %>%
    pivot_longer(
      cols = c("Active", "Fatal", "Recovered"), 
      names_to = "Status", values_to = "Total"
    ) %>%
    left_join(ausPops) %>%
    mutate(
      Status = factor(Status, levels = c("Fatal", "Recovered", "Active")),
      Rate = 1e6*Total / Population
    ) %>%
    ggplot(aes(Date, Rate, fill = Status, label = Total)) +
    geom_col() +
    geom_line(
      data = . %>%
        group_by(State, Date) %>%
        summarise(
          Rate = sum(Rate),
          Total = sum(Total)
        ) %>%
        mutate(Status = "Confirmed"),
      colour = "blue"
    ) +
    facet_wrap(~State, ncol = 4) + 
    scale_fill_manual(
      values = c(
        Active = rgb(0, 0, 0),
        Confirmed = rgb(0, 0.3, 0.7),
        Fatal = rgb(0.8, 0.2, 0.2),
        Recovered = rgb(0.2, 0.7, 0.4)
      )
    ) +
    scale_x_date(expand = expansion(c(0, 0.03))) +
    labs(y = "Rate (Cases / Million)")
)

Breakdown of individual states. Victorian recovered numbers began to be accurately reported from 22^nd March, with other states gradually providing this information. NSW/QLD recovered cases have only recently begun being reported and up until the most recent dates, recovered/active values were very approximate for these states. The extreme drop for NSW active cases in early June is a function of the changed reporting strategy implemented by NSW Health.

Daily New Cases

ggplotly(
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    group_by(State) %>%
    mutate(daily = c(0, diff(confirmed))) %>%
    ungroup() %>%
    dplyr::filter(confirmed > 0) %>%
    mutate(
      daily = case_when(
        daily < 0 ~ 0,
        daily >= 0 ~ daily
      )
    ) %>%
    bind_rows(
      group_by(., date) %>%
        summarise(daily = sum(daily)) %>%
        ungroup() %>%
        mutate(State = "All States")
    ) %>%
    group_by(State) %>%
    mutate(
      MA = round(sma(daily, 7), 2),
      MA2 = round(sma(daily, 14), 2),
      `Above Average` = MA > MA2
    ) %>%
    dplyr::filter(date > "2020-03-01") %>%
    ggplot(aes(date, daily)) +
    geom_col(
      aes(fill = `Above Average`, colour = `Above Average`),
      data = . %>% dplyr::filter(!is.na(`Above Average`)),
      width = 1/2
    ) +
    geom_line(aes(y = MA), colour = "blue") +
    geom_line(aes(y = MA2), colour = "black") +
    facet_wrap(~State, scales = "free_y") +
    labs(
      x = "Date",
      y = "Daily New Cases",
      fill = "\nAbove\nAverage"
    ) +
    scale_fill_manual(values = c("white", rgb(1, 0.2, 0.2))) +
    scale_colour_manual(values = c("grey50", rgb(1, 0.2, 0.2))),
  tooltip = c(
    "date", "daily", "MA"
  )
)

Daily new cases for each state shown against the 7-day (blue) and 14-day (black) averages. Days which the 7-day average is above the 14-day average are highlighted in red.

Australian Fatality Rate

inc <- 6
icu <- 11
d <- 7
offset <- icu + d 
minDate <- "2020-04-20"

list(
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    group_by(date) %>%
    summarise_at("confirmed", sum) %>%
    left_join(
      deaths %>%
        group_by(date) %>%
        summarise_at("deaths", sum)
    ) %>%
    dplyr::filter(
      date > minDate
    ) %>%
    mutate(
      fr = deaths / confirmed,
      type = "No Offset"
    ),
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    mutate(
      date = date + offset 
    ) %>%
    group_by(date) %>%
    summarise_at("confirmed", sum) %>%
    left_join(
      deaths %>%
        group_by(date) %>%
        summarise_at("deaths", sum) 
    ) %>%
    dplyr::filter(
      date > minDate
    ) %>%
    mutate(
      fr = deaths / confirmed,
      type = glue("Offset ({offset} days)")
    ) 
) %>%
  bind_rows() %>%
  ggplot(
    aes(date, fr, colour = type)
  ) +
  geom_line() +
  scale_x_date(
    expand = expansion(mult = 0, add = 0)
  ) +
  scale_y_continuous(label = percent) +
  labs(
    x = "Date",
    y = "Estimated Fatality Rate",
    colour = "Calculation"
  )

Fatality rate for Australian cases as calculated using two methods. Where no offset is included, the rate shown is simply the number of fatalities divided by the total number of reported cases on the same date. When cases increase during a new outbreak, this will skew the fatality rate lower. An alternative is to use an offset based on the fact the the median time from infection to symptom onset is 6 days, the median time from symptom onset to ICU admission is 11 days, and the median time from ICU admission to mortality is 7 days. When using the offset, the fatality rate is calculated as the number of recorded fatalities on a given date, divided by by the number of cases from 18 days ago. Whilst still flawed this may give a less biased estimate on the true fatality rate, and importantly, will always be higher than the alternative calculation. The intial fatality rate spiked above 30% during the intial outbreak under the offset approach, and as such, data is only shown after 20 Apr, 2020. All times used for estimation the offset were obtained from here

Current Growth Factor

n <- 14
minCases <- 1
cp <- glue(
  "*Growth factor for each State/Territory. 
  __Values are calculated using only locally-acquired cases__.
  In order to try and minimise volatility a {n} day simple moving average was used, in contrast to the 5 day average as advocated [here](https://www.abc.net.au/news/2020-04-10/coronavirus-data-australia-growth-factor-covid-19/12132478).
  This enables assessment of the growth factor over an entire quarantine period.
  This value becomes volatile when daily new cases approach zero as is commonly observed in small populations, and at the end stages of an outbreak. 
  As a result, values are only shown when the {n}-day average of new __locally acquired cases__ exceeds {minCases}.*"
)

gf <- list(
  local %>%
    dplyr::filter(date <= dt) %>%    
    arrange(date) %>%
    group_by(State) %>%
    mutate(
      new = c(0, diff(local)),
      new_ma = sma(new, n)
    ) %>%
    dplyr::filter(local > 0, !is.na(new_ma)) %>%
    mutate(
      R = c(NA, new_ma[-1] / new_ma[-n()]),
      R = case_when(
        is.nan(R) ~ NA_real_,
        new_ma < minCases ~ NA_real_,
        TRUE ~ R
      )
    ) %>%
    ungroup() %>%
    arrange(State),
  local %>%
    dplyr::filter(date <= dt) %>%
    arrange(date) %>%
    group_by(Country, date) %>%
    summarise_at(vars(local), sum) %>%
    ungroup() %>%
    mutate(
      new = c(0, diff(local)),
      new_ma = sma(new, n)
    ) %>%
    dplyr::filter(local > 0, !is.na(new_ma)) %>%
    mutate(
      R = c(NA, new_ma[-1] / new_ma[-n()]),
      R = case_when(
        is.nan(R) ~ NA_real_,
        new_ma < minCases ~ NA_real_,
        TRUE ~ R
      ),
      State = "All States"
    ) %>%
    arrange(State)
) %>%
  bind_rows() %>%
  # dplyr::filter(date > ymd("2020-03-01")) %>%
  ggplot(aes(date, R, colour = State)) +
  geom_ribbon(aes(ymin = 1, ymax = R), alpha = 0.1) +
  geom_hline(yintercept = 1) +
  geom_label(
    aes(label = R),
    data = . %>%
      dplyr::filter(date == max(date), !is.na(R)) %>%
      mutate(R = round(R, 2), date = date + 1),
    fill = rgb(1, 1, 1, 0.3),
    show.legend = FALSE,
    nudge_y = 0.3,
    size = 4
  ) +
  labs(
    x = "Date", y = "Growth Factor"
  ) +
  facet_wrap(~State, scales = "free_x") +
  theme(legend.position = "none") +
  coord_cartesian(ylim = c(0.5, 1.8))#2.1))
gf

Growth factor for each State/Territory. Values are calculated using only locally-acquired cases. In order to try and minimise volatility a 14 day simple moving average was used, in contrast to the 5 day average as advocated here. This enables assessment of the growth factor over an entire quarantine period. This value becomes volatile when daily new cases approach zero as is commonly observed in small populations, and at the end stages of an outbreak. As a result, values are only shown when the 14-day average of new locally acquired cases exceeds 1.

The current 14 day growth factor is 1.02 which gives considerable cause for concern..

Testing Within Each State

tested %>% 
  left_join(confirmed, by = c("State", "Country", "date") ) %>%
  dplyr::filter(date == dt) %>%
  left_join(ausPops,  by = "State") %>%
  bind_rows(
    tibble(
      State = "National Total",
      date = dt,
      Population = sum(.$Population, na.rm = TRUE),
      confirmed = sum(.$confirmed, na.rm = TRUE),
      tests = sum(.$tests, na.rm = TRUE)
    )
  ) %>%
  mutate(
    `Tests / '000` = round(1e3 * tests / Population, 2),
    Positive = confirmed / tests,
    Negative = 1 - Positive,
    isTotal = grepl("Total", State)
  ) %>%
  dplyr::select(
    State, Population,
    Confirmed = confirmed,
    Tests = tests, 
    contains("000"), 
    ends_with("ive"),
    isTotal
  ) %>%
  arrange(isTotal, desc(`Tests / '000`)) %>%
  dplyr::select(-isTotal) %>%
  dplyr::rename(
    `% Positive Tests` = Positive,
    `% Negative Tests` = Negative
  ) %>%
  mutate_at(
    vars(starts_with("%")), percent, accuracy = 0.01
  ) %>%
  pander(
    justify = "lrrrrrr",
    missing = "",
    caption = glue(
      "*COVID-19 testing scaled by state population size.
      Confirmed cases are assumed to be the tests returning a positive result.
      The current numbers available for some states are a lower limit, and as such, the proportion of the population tested is likely to be higher, as is the proportion of tests returning a negative result.*"
    ),
    emphasize.strong.rows = nrow(.)
  )

COVID-19 testing scaled by state population size. Confirmed cases are assumed to be the tests returning a positive result. The current numbers available for some states are a lower limit, and as such, the proportion of the population tested is likely to be higher, as is the proportion of tests returning a negative result.

State	Population	Confirmed	Tests	Tests / '000	% Positive Tests	% Negative Tests
Victoria	6,629,870	20,411	4,176,707	630	0.49%	99.51%
New South Wales	8,117,976	5,034	4,388,407	540.6	0.11%	99.89%
South Australia	1,756,494	588	893,844	508.9	0.07%	99.93%
Northern Territory	245,562	90	88,324	359.7	0.10%	99.90%
Australian Capital Territory	428,060	118	148,660	347.3	0.08%	99.92%
Queensland	5,115,451	1,281	1,621,595	317	0.08%	99.92%
Tasmania	535,500	234	150,329	280.7	0.16%	99.84%
Western Australia	2,630,557	878	670,126	254.8	0.13%	99.87%
National Total	25,459,470	28,634	12,137,992	476.8	0.24%	99.76%

R Session Information

R version 4.0.3 (2020-10-10)

**Platform:** x86_64-pc-linux-gnu (64-bit)

locale: LC_CTYPE=C, LC_NUMERIC=C, LC_TIME=C, LC_COLLATE=C, LC_MONETARY=C, LC_MESSAGES=en_AU.UTF-8, LC_PAPER=en_AU.UTF-8, LC_NAME=C, LC_ADDRESS=C, LC_TELEPHONE=C, LC_MEASUREMENT=en_AU.UTF-8 and LC_IDENTIFICATION=C

attached base packages: stats, graphics, grDevices, utils, datasets, methods and base

other attached packages: jsonlite(v.1.7.2), QuantTools(v.0.5.7.1), data.table(v.1.13.6), plotly(v.4.9.3), pander(v.0.6.3), rvest(v.0.3.6), xml2(v.1.3.2), glue(v.1.4.2), ggrepel(v.0.9.0), scales(v.1.1.1), lubridate(v.1.7.9.2), rlang(v.0.4.10), forcats(v.0.5.0), stringr(v.1.4.0), dplyr(v.1.0.2), purrr(v.0.3.4), readr(v.1.4.0), tidyr(v.1.1.2), tibble(v.3.0.4), ggplot2(v.3.3.3) and tidyverse(v.1.3.0)

loaded via a namespace (and not attached): Rcpp(v.1.0.5), ps(v.1.5.0), assertthat(v.0.2.1), digest(v.0.6.27), R6(v.2.5.0), cellranger(v.1.1.0), backports(v.1.2.1), reprex(v.0.3.0), evaluate(v.0.14), highr(v.0.8), httr(v.1.4.2), pillar(v.1.4.7), lazyeval(v.0.2.2), curl(v.4.3), readxl(v.1.3.1), rstudioapi(v.0.13), rmarkdown(v.2.6), labeling(v.0.4.2), selectr(v.0.4-2), htmlwidgets(v.1.5.3), munsell(v.0.5.0), broom(v.0.7.3), compiler(v.4.0.3), modelr(v.0.1.8), xfun(v.0.20), pkgconfig(v.2.0.3), htmltools(v.0.5.0), tidyselect(v.1.1.0), fasttime(v.1.0-2), fansi(v.0.4.1), viridisLite(v.0.3.0), crayon(v.1.3.4), dbplyr(v.2.0.0), withr(v.2.3.0), grid(v.4.0.3), gtable(v.0.3.0), lifecycle(v.0.2.0), DBI(v.1.1.0), magrittr(v.2.0.1), cli(v.2.2.0), stringi(v.1.5.3), farver(v.2.0.3), fs(v.1.5.0), ellipsis(v.0.3.1), generics(v.0.1.0), vctrs(v.0.3.6), tools(v.4.0.3), crosstalk(v.1.1.0.1), hms(v.0.5.3), yaml(v.2.2.1), colorspace(v.2.0-0), knitr(v.1.30) and haven(v.2.3.1)